Gabriela Csurka
@gabrielacsurka.bsky.social
Principal research scientist at Naver Labs Europe, I am interested in most aspects of computer vision, including 3D scene reconstruction and understanding, visual localization, image-text joint representation, embodied AI, ...
Reposted by Gabriela Csurka
For which the code is also available github.com/naver/pow3r
GitHub - naver/pow3r
Contribute to naver/pow3r development by creating an account on GitHub.
github.com
June 12, 2025 at 1:41 PM
For which the code is also available github.com/naver/pow3r
Reposted by Gabriela Csurka
1/ 📄 Paper 1:
"DUNE: Distilling a UNiversal Encoder from Heterogeneous 2D and 3D Teachers"
We propose DUNE: a ViT-based encoder distilled from multiple specialized 2D & 3D foundation models to unify visual tasks across 2D, 3D and human understanding.
"DUNE: Distilling a UNiversal Encoder from Heterogeneous 2D and 3D Teachers"
We propose DUNE: a ViT-based encoder distilled from multiple specialized 2D & 3D foundation models to unify visual tasks across 2D, 3D and human understanding.
June 9, 2025 at 11:07 AM
1/ 📄 Paper 1:
"DUNE: Distilling a UNiversal Encoder from Heterogeneous 2D and 3D Teachers"
We propose DUNE: a ViT-based encoder distilled from multiple specialized 2D & 3D foundation models to unify visual tasks across 2D, 3D and human understanding.
"DUNE: Distilling a UNiversal Encoder from Heterogeneous 2D and 3D Teachers"
We propose DUNE: a ViT-based encoder distilled from multiple specialized 2D & 3D foundation models to unify visual tasks across 2D, 3D and human understanding.
Reposted by Gabriela Csurka
Interestingly, the network generalizes very well to many more images than seen during training.
Code IS available here github.com/naver/must3r
I hope it works in your scenarios and you have as much fun as we do playing around with it!
Code IS available here github.com/naver/must3r
I hope it works in your scenarios and you have as much fun as we do playing around with it!
GitHub - naver/must3r: MUSt3R: Multi-view Network for Stereo 3D Reconstruction
MUSt3R: Multi-view Network for Stereo 3D Reconstruction - naver/must3r
github.com
June 12, 2025 at 12:35 PM
Interestingly, the network generalizes very well to many more images than seen during training.
Code IS available here github.com/naver/must3r
I hope it works in your scenarios and you have as much fun as we do playing around with it!
Code IS available here github.com/naver/must3r
I hope it works in your scenarios and you have as much fun as we do playing around with it!